3826
Biochemistry 2003, 42, 3826-3834
Mechanistic Insights into Replication Across from Bulky DNA Adducts: A Mutant Polymerase I Allows an N-Acetyl-2-aminofluorene Adduct To Be Accommodated during DNA Synthesis† Samer Lone and Louis J. Romano* Department of Chemistry, Wayne State UniVersity, Detroit, Michigan 48202 ReceiVed December 4, 2002; ReVised Manuscript ReceiVed February 5, 2003
ABSTRACT: The molecular mechanism that allows a polymerase to incorporate a nucleotide opposite a DNA lesion is not well-understood. One way to study this process is to characterize the altered molecular interactions that occur between the polymerase and a damaged template. Prior studies have determined the polymerase-template dissociation constants and used kinetic analyses and a protease digestion assay to measure the effect of various DNA adducts positioned in the active site of Klenow fragment (KF). Here, a mutator polymerase was used in which the tyrosine at position 766 of the KF has been replaced with a serine. This position is located at the junction of the fingers and palm domain and is thought to be involved in maintaining the active site geometry. The primer-template was modified with N-acetyl-2aminofluorene (AAF), a well-studied carcinogenic adduct. The mutant polymerase displayed a significant increase in the rate of incorporation of the correct nucleotide opposite the adduct but was much less prone to incorporate an incorrect nucleotide relative to the wild-type polymerase. Both the wild-type and the mutant polymerase bound much more tightly to the AAF-modified primer-template; however, unlike the wild-type polymerase, the binding strength of the mutant was influenced by the presence of a dNTP. Moreover, the mutant polymerase was able to undergo a dNTP-induced conformational change when the AAF adduct was positioned in the active site, while the wild-type enzyme could not. A model is proposed in which the looser active site of the mutant is able to better accommodate the AAF adduct.
The mechanism used by DNA polymerases to accurately incorporate a nucleotide during DNA replication has been the subject of intense study for the past four decades (1). Kinetics and crystallographic analyses have shown that this process proceeds through a number of sequential steps that involve at least two different conformations of the DNA polymerase (2-4). The conformational rearrangement from an open binary to a closed ternary complex that is induced by the binding of a dNTP results in a structure in which the dNTP is paired with the template base and is properly aligned for the nucleophilic attack by the 3′-hydroxyl of the primer. Apparently, polymerases are only able to reach a fully active catalytic configuration if the nucleotide base can adopt a Watson-Crick geometry with the template base, and it is thought that this process provides the much of the selectivity during nucleotide incorporation. The similarity of the mechanistic details among a variety of polymerases is likely the result of the remarkable resemblance of the structures of these enzymes. The general structure of the polymerase has been compared to that of a human right-hand with three distinct subdomains designated as the palm, fingers, and thumb. In this model, the palm subdomain contains highly conserved catalytic residues that are involved in the phosphoryl transfer reaction, the fingers † This investigation was supported by Public Health Service Grant CA40605 awarded by the Department of Health and Human Services. * To whom correspondence should be addressed. Tel.: (313) 5772584. Fax: (313) 577-8822. E-mail:
[email protected].
subdomain interacts with both the incoming nucleotide and the template base it will be paired with, and the thumb subdomain is involved in positioning the duplex DNA and has been shown to have roles in both processivity and translocation (reviewed in refs 4-6). The fingers subdomain contains several planer hydrophobic amino acid residues that participate in the correct positioning of the dNTP when the closed conformation is formed (3, 7, 8). One of the best studied of these residues is the tyrosine at position 766, which is located at the base of the O helix in the fingers subdomain near the junction with the palm (5, 9). In both the Taq and the Bst polymerase open complexes, this tyrosine side chain is stacked above the terminal base pair of the primer-template. In the closed complex, the tyrosine has rotated away from the DNA, and the template base has moved into the position it had occupied allowing the formation of the base pair with the incoming dNTP (3, 10). Studies with DNA polymerase I (Klenow fragment) (KF)1 have shown that Y766S and Y766A mutants have mutator phenotypes (11), and analogous mutations in eukaryotic polymerase β also lead to higher levels of mutagenesis (12, 13). It is likely that the reduced fidelity of these mutants is caused by a different active site conforma1 Abbreviations: AAF, N-acetyl-2-aminofluorene; AF, 2-aminofluorene; KF, Klenow fragment; dG-C8-AAF, N-(2′-deoxyguanosin8-yl)-2-acetylaminofluorene; dG-C8-AF, N-(2′-deoxyguanosin-8-yl)2-aminofluorene; PAGE, polyacrylamide gel electrophoresis; KM, Michaelis constant; Vmax, the maximum rate of reaction; Fins, frequency of insertion; Fext, frequency of extension; Kd, dissociation constant.
10.1021/bi027297t CCC: $25.00 © 2003 American Chemical Society Published on Web 03/15/2003
Mutant Polymerase Accommodates Bulky DNA Adduct
FIGURE 1: Structure of the N-(deoxyguanosin-8-yl)-2-acetylaminofluorene adduct (dG-C8-AAF).
tion, possibly resulting from an alteration in the conversion from the open to the closed form. Although a great deal is known regarding the mechanism of synthesis by a polymerase using a normal DNA template, very little is known about how a bulky DNA lesion affects this process. An understanding of how a polymerase interacts with this class of DNA damage is particularly important because of the discovery of the bypass or Y-family polymerases, whose role is to carry out DNA synthesis past DNA damage that otherwise blocks replication (14, 15). There have been a few indirect studies that have attempted to explain the effects of bulky lesions on DNA replication by correlating the structures of these adducts in double-stranded DNA or at a primer-template junction with the mutagenic event that a particular structure induces (16, 17). We have developed several methods to study directly the interactions between a polymerase and a primer-template and have used them to measure how the presence of bulky adducts in the active site affects these interactions. In this regard, we have used a gel retardation assay (18, 19) and a limited trypsin digestion analysis (20) to show that bulky lesions, such as benzo[a]pyrene, 2-aminofluorene (AF), and N-acetyl-2-aminofluorene (AAF) adducts can be well-accommodated within the active site of the polymerase in the open binary complex but that these adducts can interfere with the conformational change of the polymerase to the closed catalytically active ternary complex. Presumably, these bulky adducts alter the structure and stability of the ternary complex, thus affecting an important step that is crucial in determining the fidelity of the nucleotide incorporation step (1). One of the best studied bulky DNA adduct is one formed by treatment of cells with the potent model chemical carcinogen, AAF. Two major adducts result from this exposure: the N-(2′-deoxyguanosin-8-yl)-2-acetylaminofluorene adduct (dG-C8-AAF) (Figure 1) and its deacetylated derivative, the N-(2′-deoxyguanosin-8-yl)-2-aminofluorene adduct (dG-C8-AF). Although the AF adduct has been shown to be easily bypassed during in vitro DNA synthesis, the AAF adduct represents a strong block to synthesis by all replicative polymerases studied (21, 22). These differences are thought to be related to the structure that each adduct presents to the DNA polymerase. Most spectral, enzymatic, and theoretical studies suggest that the AF structure produces much less distortion in the DNA helix than the AAF adduct
Biochemistry, Vol. 42, No. 13, 2003 3827 (23-27). Multidimensional NMR experiments have shown that the guanine bearing the C8-AAF adduct rotates from anti to syn conformation in double-stranded DNA helix with the fluorene ring inserted into the helix (base displacement model, ref 28). In the present study, we report our initial efforts aimed at understanding how specific amino acid residues in the polymerase active site affect the interactions with a modified DNA template. We find that substitution of the tyrosine at position 766 of KF with a serine substantially alters the interactions and mechanism of action of the polymerase on both unmodified and AAF-modified DNA. Not only does this mutation affect the rate of nucleotide incorporation opposite the adduct, but it also alters the effect of dNTPs on both the binding to the primer-template and the conformational change to the closed ternary complex. MATERIALS AND METHODS Materials. Wild-type and Y766S Klenow fragment (exo-) clones were generously provided by Dr. Catherine Joyce of Yale University. Both WT and Y766S Proteins were overexpressed and purified as described (29) and contained the D424A mutation, which eliminates the 3′-5′ exonuclease activity. The specific activity of the protein was determined as described (30). Protein concentrations were determined colorimetrically by the Bradford assay (31) using Bio-Rad laboratory reagents. T4 polynucleotide kinase was purchased from Amersham Pharmacia Biotech. Trypsin and terminal deoxynucleotide transferase came from Boehringer Mannheim. Oligonucleotides were obtained from Midland Certified Inc. dNTPs were purchased from Amersham Pharmacia Biotech. [γ-32P]ATP was from ICN Biomedicals. Synthesis and Purification of Oligonucleotides. All oligonucleotides were purified by denaturing polyacrylamide gel electrophoresis. Site-specifically modified 28-mer templates were synthesized, purified, and characterized as described (18). The primers lacking 3′-OH were obtained by extension of corresponding oligonucleotides with ddNMPs using terminal deoxynucleotide transferase as described (20). Primer Extension Analysis. 32P-labeled 12-mer primer (1 nM) was annealed to a 2-fold excess of the AAF-modified 28mer template (2 nM). The reactions were started with dNTPs (100 µM) and 10 nM KF in 50 mM Tris-HCl, pH 7.5, containing 10 mM MgCl2, 1 mM dithiothreitol, and 0.05 mg/mL bovine serum albumin. At the indicated time points after the addition of the polymerase (0-60 min), 5 µL aliquots of the reaction mixture were taken, and the reaction was stopped by addition to 10 µL of gel loading buffer containing 90% formamide and 5 mg/mL bromophenol blue and xylene cyanol. The samples were analyzed on a 20% denaturing polyacrylamide gel. Product formation was measured by phosphoimager analysis and quantified using Molecular Dynamics ImageQuant. Total bypass synthesis was determined by dividing the total radioactivity across from and extended past the adduct at each time point by the total radioactivity in each lane. Full extension was determined by dividing the total radioactivity of the 28-mer product by the total radioactivity in each lane. Steady-State Kinetics. Steady-state kinetics using standingstart single nucleotide insertions and extensions were carried
3828 Biochemistry, Vol. 42, No. 13, 2003 out similarly to those described (32). Typical reactions were carried out in 10 µL volumes in the presence of 50 mM TrisHCl, pH 7.5, 10 mM MgCl2, 1 mM DTT, and 0.05 mg/mL BSA. 0.1 µM primer-templates were annealed by heating to 90° and slow cooling. Reaction mixtures containing 0.0010.03 units of DNA polymerase were incubated at room temperature for 1-12 min. Both polymerase concentrations and times were varied for each nucleotide examined so that less than 20% incorporation occurred. The extent of each reaction was determined by running quenched reaction samples (95% formamide, 20 mM EDTA, 0.05% xylene cylanol, and bromophenol blue) on a 20% denaturing polyacrylamide gel to separate unreacted primer from insertion products. As described previously (33), relative velocities were calculated as the extent of the reaction divided by the reaction time and normalized for the varying enzyme concentrations used. The Michaelis constant (Km) and maximum rate of the reaction (Vmax) were obtained from Hanes-Wolf plots of the kinetic data. Insertion (Fins) and extension (Fext) frequencies were determined relative to dC: dG and dA:dT, respectively, according to equations developed by Mendelman et al. (34, 35). The frequency of insertion and extension are defined as F ) (Vmax/Km)[wrong pair]/(Vmax/Km)[right pair]. All reactions reported represent an average of at least three experiments and had standard deviations less than 20%. Gel Retardation Assay. Equilibrium dissociation constants (Kd) for the polymerase-primer-template complexes were determined as described (18). Increasing amounts of KF (typically 0-200 nM) were incubated with 32P-labeled primer-templates (5-50 pM) for 30 min at room temperature in 50 mM Tris-HCl, pH 7.5, containing 10 mM MgCl2, 1 mM dithiothreitol, 0.05 mg/mL bovine serum albumin, 4% glycerol, and 0.4 mM dNTP (if present). The reaction mixtures were analyzed on a native 7% polyacrylamide gel preequilibrated with 36 mM Tris borate buffer, pH 8.3. Quantification was performed using Molecular Dynamics PhosphorImager and ImageQuant. The amount of proteinDNA complex formed at equilibrium was calculated as the difference in the band intensities of the initial primertemplates without polymerase addition and unbound primertemplates. To determine Kd, the fraction of the DNA bound to the protein was plotted against the initial protein concentrations, and the data was analyzed using Ultrafit (Biosoft, Cambridge, UK) by fitting to the equation for single-site ligand binding. Each determination represents the average of at least three independent experiments. Tryptic Digestion of KF Bound to Unmodified and AAFModified Primer-Templates. The polymerase-DNA complexes were formed in 50 mM Tris-HCl, pH 7.5, containing 10 mM MgCl2 and 1 mM dithiothreitol. The binding was carried out at room temperature for 15 min in a 12 µL reaction containing 0.6 µM annealed primer-template, 0.3 µM KF (exo-), and 0-16 mM dNTP. Two microliters of trypsin solution in water (15 ug/mL final) was added to each reaction mixture, and the digestion was terminated after 6 s by addition of 6 µL of SDS sample buffer containing 0.125 M Tris-HCl, pH 6.8, 6% SDS, 30% glycerol, and 10 µg/mL bromophenol blue. The samples were loaded on a 10% SDS gel, and the electrophoresis was performed according to standard procedure (36). Gels were fixed and stained using the Silver Stain Plus Kit (BIO-RAD) according to the
Lone and Romano manufacturer’s protocol. At 16 mM dNTP concentration, the relative levels of cleavage were determined by scanning the gels and using NIH Image to quantify the intensity of each band. Each determination is representative of two to three individual experiments. RESULTS Primer Extensions using the AAF-Modified Templates. As a first step to determine the effect of the replacement of tyrosine with serine at position 766 of KF on the replication of AAF-modified DNA, DNA synthesis was carried out using the primer-template shown in Figure 2. Using wildtype KF, most of the DNA synthesis on this template was stalled at a position one nucleotide before the AAF adduct in the template. Over the 60 min time course, no more than 20% of the synthesis occurred across from the adduct or extended past the adduct position (Figure 2A,B). With the Y766S mutant polymerase, synthesis was also blocked one nucleotide before the adduct position, but in this case a much higher percentage of synthesis was able to occur across from the adduct (Figure 2A): after 60 min, approximately 40% of the product extended to the position across from the AAF adduct (Figure 2B). Interestingly, a much smaller percentage of full extension occurred, using this mutant (Figure 2C), whereas the wild-type KF was able to give 6% full extension, and the mutant only produced about 1%. Thus, while the mutant was better able to incorporate across from the adduct, it was less able to extend from this structure. Single Nucleotide Incorporation Kinetics. To more fully explore the differences between the wild-type and the Y766S polymerases, a single nucleotide steady-state kinetic analysis was carried out using the templates shown at the top of Table 1. Using the unmodified primer-template, both the wild-type KF and the Y766S mutant had a strong preference for the incorporation of dC across from the template dG, with the Vmax/Km for Y766S about 1/3 that of the wild-type KF. Thus, the Fins for the incorporation of dG, dA, and dT across from a dG was 3.6, 12, and 72 times greater, respectively, for the mutant than was found for wild-type KF (Table 1), although the differences in the absolute values of Vmax/Km for the two polymerases for the incorporation of the incorrect nucleotides were smaller (Table 1). Prior studies compared the steady-state rate of misinsertion of dTTP by this mutant across from an unmodified dG (37). Both this study and the data shown in Table 1 indicate that the mutant enzyme is able to incorporate dTTP across from a dG about 50-fold faster. When these experiments were repeated with the AAFmodified primer-templates, the Vmax/Km for the incorporation of dC across from the modified G was 16 times greater for the mutant polymerase (Table 1), confirming the enhanced ability of Y766S to incorporate a nucleotide opposite the adduct as shown in Figure 2. Interestingly, the reduced ability of the mutant to discriminate between the correct and the incorrect nucleotide that was observed in the case of the unmodified template was not observed using an AAFmodified template. Although the absolute rate of misincorporation by the mutant is equal to or higher than that observed for the wild-type polymerase (compare Vmax/Km values in Table 1), the Fins is lower for the Y766S mutant for the incorporation of each of the incorrect nucleotides.
Mutant Polymerase Accommodates Bulky DNA Adduct
Biochemistry, Vol. 42, No. 13, 2003 3829
FIGURE 2: (A) Running-start primer extension analysis of AAF-modified primer-templates. 12-mer primer was annealed to the 28-mer template, containing an adduct at the position indicated. For both WT and Y766S Klenow fragments (exo-) all four dNTPs were added, and the reaction was terminated after the period of time indicated under each lane. (B) The insertion and extension products were quantified as explained in Materials and Methods and were plotted as a function of time. Table 1: Kinetic Parameters for Insertion by Wild-Type and Y766S Klenow Fragmentsa
WT
Y766S
dNTP:Xb
Vmax/Km (% min-1 µM-1)
Fcins
dCTP:dG dGTP:dG dATP:dG dTTP:dG dCTP:dG-AAF dGTP:dG-AAF dATP:dG-AAF dTTP:dG-AAF
22.6 3.0 × 10-2 1.3 × 10-2 8.2 × 10-3 4.4 × 10-2 3.5 × 10-4 6.9 × 10-3 1.6 × 10-3
1 1.3 × 10-3 5.8 × 10-4 3.6 × 10-4 1 8.0 × 10-3 1.6 × 10-1 3.6 × 10-2
Vmax/Km (% min-1 µM-1) 6.6 3.1 × 10-2 4.5 × 10-2 1.7 × 10-1 7.3 × 10-1 4.3 × 10-4 6.7 × 10-3 4.7 × 10-3
Fins
Fins (Y766s)/ Fins (WT)
1 4.7 × 10-3 6.8 × 10-3 2.6 × 10-2 1 5.9 × 10-4 9.2 × 10-3 6.4 × 10-3
1 3.6 12 72 1 0.074 0.058 0.18
a Kinetics of insertion were determined as described under Materials and Methods using the primer-template shown above. All values represent the mean of at least three experiments and have standard deviations
